EXAMINING THE EVIDENCE A careful review of recent data, however, indicates some important inconsistencies in the relationship between crowding and health which throw some doubt on this generally accepted formulation, particularly on the processes through which crowding may influence health. It would appear from these data that the relationship between crowding and health is a far more complex phenomenon than was originally envisaged; while under certain circumstances crowding is clearly associated with poor health states, under other circumstances it may be neutral or even beneficial. The data which cast some doubt on this relationship will be briefly reviewed under the same categories that have provided the evidence used to support the notion that crowding inevitably leads to deleterious health consequences. Urban-Rural Death and Morbidity Rates As is shown in Figure 1, death rates for all causes in the United States were indeed higher in urban areas than in rural before 1950. By 1960, however, the ratio had reversed; rural rates were higher than urban, and since 1960 the ratio of rural to urban deaths has been steadily increasing.* Paradoxically, URBAN RURAL URBAN RURAL URBAN RURAL URBAN RURAL Figure 1. Death rates from all causes by place of residence, United States, 1940, 1950,1960, 1966. Sources: (2-7). *The magnitude of these ratios should be accepted as only approximate, especially the data for 1966 which were based on population projections. Further, the differing definitions of "urban" used by the Bureau of the Census (from which the population figures were drawn) and by the National Office of Vital Statistics may introduce a source of bias.o insults such as drugs, microorganisms, and X rays is reduced.